Weather derivatives trading is undergoing exponential growth. The current size of the weather derivatives market is $8bn, and it is expected to grow to at least $300bn dollars within a few years. The opportunity to reduce exposure to natural risks will be exploited by large corporate groups and, eventually, smaller players too. It is important for the future development of the weather market that players have confidence in their own ability to determine fair prices, and that of the market to deliver these. Such confidence will encourage liquidity in the market from the end-users. However, traditional derivatives exchanges may find it difficult to obtain volume (and hence liquidity) in weather markets.

A classic example given to illustrate the benefits of portfolio diversification involves an investor allocating funds between an ice-cream vendor and an umbrella shop. If it is hot and sunny, only investment in the ice cream vendor is likely to secure a profit. If it is cold and rainy, the umbrella shop alone should generate profits. Investment diversified between the two will allow the investor to hedge against the natural risks, and ensure some level of profits under either set of conditions. The weather derivatives market allows the shop-owners to hedge their own risks without allocating funds to invest in the other's shop.

Every individual business has its own unique exposure to weather. Weather in the context of the weather risk market refers to daily fluctuations of factors such as temperatures and precipitation (rainfall). As such, the market covers deviations from the daily average. The impact of daily weather conditions is analysed over a period of time, usually measured in months. Deals can be structured to express the variation of those parameters which best address whatever risk an individual organisation is facing.

Market instruments
A weather derivative is a financial instrument that has a pay-off derived from one or more independently measurable weather parameters. These can include temperature, rainfall, wind speed, snow depth or hours of sunshine as recorded at one or more specified reference locations. Transactions are customisable in terms of the reference location, the time period of the transaction and the weather variable, level of precipitation, temperature, etc.

At present, the majority of weather deals are swaps or options based on underlying temperature indices at single locations. However, it is possible to structure hybrid deals with payouts based on a combination of two or more weather ‘underlyings'. For example, a derivative bought by a hydroelectric power generator might be based on both precipitation and temperature. Such a deal protects the buyer from both supply-side and demand-side volume risks. Hedging against low rainfall covers the risk of insufficient electricity generation, while hedging against warm weather covers the risk of insufficient demand.

The most common tool to facilitate trading in temperature outcomes is the use of ‘degree-days.' A degree-day is a measure of the variation of one day's temperature against a standard reference temperature. If the actual temperature is lower than the reference, then heating is required to achieve the reference level. The need for one degree of heat on one day is called a heating degree-day (HDD). If the reference level is set at 18ºC and the actual temperature over three days is 16ºC, 15 ºC and 19ºC, the outcome is 5HDD. Two degrees of heating are required on the first day, three on the second and none on the third. Negative values are not possible, so the third value requires 0 rather than –1HDD. Similarly, cooling degree-days (CDD) measure the number of degrees which must be taken off the actual temperature to achieve the reference level. The same pattern of temperatures would produce a value of 1CDD.

Historically, the reference temperature has been set at 65ºF/18ºC. This is the baseline relevant to the US energy industry because of consumer energy consumption behaviour for heating and cooling. However, in other industries, baselines and indeed measures should be those pertinent to the risks faced by the particular industry. For example, agriculture-based transactions can be designed to trigger when there are more than a certain number of days with minimum temperatures in the freezing range over a specific period of time.

Most trading occurs around two main seasons, winter and summer. Winter is the most significant in terms of the amount of risk to be hedged, and activity is centred around this season.

The weather market has been predominantly focused so far on temperature risk, covering variations in cold versus warm summers and cold versus warm winters. A typical temperature-derived transaction may involve a company taking out a 30-day CDD swap with a reference temperature of 65ºF. If the average temperature on each day is 70ºF degrees fahrenheit, the company is due 150 (30x5) degree-days multiplied by the sum of money agreed for each degree-day. If the company had taken out a HDD swap, it would have owed the same amount of money. Degree-day instruments, whether as swaps or options, are the standard weather derivatives used at the present time.

However, the market is evolving to deal with other types of weather condition:

  • Precipitation risk. A similar approach to degree-days could be used for precipitation. By looking at cumulative precipitation over a period of months, deals can be structured so pay-outs are based on whether there is cumulative rainfall above or below certain thresholds. For example, a ski resort which wants to hedge the risk of having a bad winter in terms of low snowfall can determine a threshold in terms of what level of snow would be bad for business. A transaction can then be structured that will cover the business if snow falls below that level.
  • Stream flow risk. More complex composites such as stream flow risk are also being considered. The chances of flooding in streams, or levels falling very low, are often more directly important to businesses than actual precipitation levels. The difficulty lies in using science to predict stream flow risks, and then applying a suitable model to price the market.

    The structure of transactions could also be based on what is called the ‘misery index', which combines temperature and humidity. A theme park, by correlating its historical daily admissions statistics against the historical weather conditions, might identify that admission rates drop substantially when the misery index is very high because nobody wants to go out in that weather. It could then hedge against that measure of weather.

    Market development
    The weather derivatives market began in 1997 in the US. It has since grown from being used predominantly to cover the winter season exposure to cover the summer season as well.

    Options products typically traded between wholesale counterparties have been the norm, although end-users have started coming into the market. There is no hard source of data on numbers since only a small proportion of transactions are carried out in standard products. Most involve tailor-made over-the-counter (OTC) deals, which are by definition executed outside exchanges.

    It is estimated that around 4,000 deals have already been transacted in the US. At present, typical deal sizes are between $2m and $30m. Brokers involved in the market include United, TFS, Euro Brokers, Prebon and Nat Source. As well as the OTC market, weather derivatives are traded round the clock on the Chicago Mercantile Exchange (CME). The CME launched the first exchange-traded weather derivatives on 22 September 1999, and expanded its geographic coverage significantly in early 2000.

    There are two principal markets outside of the US:

  • Europe. The first European deal was struck in 1998. The European market, centred on the UK, is rapidly catching up with the US. The first trade in Europe was a swap transacted between Enron and Scottish Hydro. A further seven trades went through during 1999. These trades included a swap between ENW and Enron, a collar between Soccram and Société Générale and five other option trades. It is now believed that there have now been about 40-50 trades, of which 80% are swaps. The total notional volume is about £20m ($28m), or approximately 1.5% of that in the US. Volume within the UK market has been focused on degree-days based on 18ºC.
  • Japan. The first trade in Japan occurred in June last year between Société Générale and an end user. Settlement and liquidity within the weather derivative market in Japan have been aided with weather data being available on Reuters. There is also movement at the international level., an internet-based weather derivatives exchange, opens for trading in the third quarter of this year. It is backed by LIFFE, the London International Financial Futures and Options Exchange, and will trade in ‘semi-standardised' contracts.

    Problems and prospects
    Most trade occurs in OTC products rather than standard exchange-settled derivatives. OTC markets currently suffer from difficulties in establishing accepted pricing models for weather risk contracts.

    This problem is compounded by a lack of transparency which makes price discovery for weather contracts even more difficult. These problems can lead to very large spreads in OTC contracts. It also creates uncertainty from the buyers' perspective as to whether or not they are receiving a fair arrangement from their OTC counter party.

    Improved information flows should help counter some of these problems. Official bodies such as the UK Meteorological Office are working with industry players to create a fair pricing policy on data sales. However, it will be harder to dislodge concerns about market liquidity and hence fair and competitive pricing. Although there are potentially great gains to be had through further development of the market, this may hinge on the attitudes and actions of major established players. By their nature, weather derivatives will tend to attract organisations that, at least in these early stages, are not familiar with or do not have access to current exchange-based markets. The creation of standardised products will be of benefit to end-users because prices can be measured against the standardised market price. Although the larger players may prefer now to trade in tailored OTC products, efficiency gains may be had from encouraging market development.

  • Topics